Integrated tool for automatically generating probability based, nat ionally and state representative samples across the globe

Abstract

This disclosure generally relates to a novel tool that has the potential to be deployed during public health emergencies and used by professionals with little to no expertise in survey methodology. Because sample sizes are estimated correctly and sampling executed speedily and validly, there can be more confident in the generated data for evidence-based decision-making.

The invention claimed is: 
     
         1 . An algorithm for a web-based tool to streamline sample size and sampling operations to yield representative samples for individual countries around the globe wherein:
 The system as per claim  1 , calculates the sample size needed for the survey based on the prevalence of the outcome;   The system as per claim  1 , estimates the number of counties that could feasibly be accessed based on the user's self-reported budget and mode of survey administration,   The system as per claim  1 , probabilistically selects which counties have to be included in the survey, and the number of people that should be surveyed in each of the chosen counties. All the user must do thereafter is to use the output from the application to field the survey accordingly.   
     
     
         2 . A survey tool that includes components:
 As per claim  2 , the information gathering component that collects information from the user on the particulars of their survey, including the key outcome, the country where it is being conducted, their budget for the total survey (will determine how many counties will be sampled), and the mode of survey administration;   As per claim  2 , the Publicly available records from past and ongoing surveys in the U.S. (obtained from Parts A and B of regulatory submissions sent to the Office on Management and Budget for approval as required by the paperwork Reduction Act), will be used to estimate the incremental cost of adding one more county to a national survey (using a linear regression model);   As per claim  2 , an estimate will serve as a benchmark to estimate how many counties a given budget could support, considering the mode of data collection (e.g., face to face, or telephone), which have different associated implementation costs in the real world;   As per claim  2 , a large database containing prevalence estimates for hundreds of different outcomes for all countries of the world separately;   As per claim  2 , the database will be curated from authoritative sources of information such as ‘Our World in Data’; official reports from the World Health Organization, the United Nations, and federal Government reports; scientific articles; white papers; and several other reviewed sources;   As per claim  2 , the database will provide the prevalence estimates to be input for sample size calculation for each country of interest;   As per claim  2 , the large database comprises an exhaustive list of all the counties or county-equivalents for all countries of the world, along with their most recent census estimates or projections.   
     
     
         3 . An integrated tool for automatically generating probability-based, nationally and state-representative samples across the globe where the output includes:
 As per claim  3 , the selected counties or county-equivalents;   As per claim  3 , the population of each of the counties selected;   As per claim  3 , the probability of selection at the first stage of sampling (i.e., the primary sampling unit, or the county level);   As per claim  3 , the number of people to be selected in each county that was sampled;   As per claim  3 , the probability of selection at the second stage of sampling (i.e., individual-level sampling);   As per claim  3 , the joint probability of selection accounts for both the first and second stages of sampling.